The MHRA has issued guidelines aimed at sponsors intending to use real-word data (RWD) sources in prospective randomised clinical trials (RCTs). The guidelines form part of a series from the MHRA on the use of RWD to support regulatory decision-making (for an introduction to the MHRA's latest guidance on data quality and RWD, please click here). 

We've set out 6 points to note on the guidelines below.

  1. What do the guidelines cover? The guidelines apply to both clinical trial authorisation (where a trial will be run at least partly in the UK) and clinical trial design. The MHRA anticipates that randomised trials using a RWD source may be most commonly used to facilitate label changes for products that have already been approved, including drug repurposing. However, the MHRA points out that it is at least possible that RWD may be used in support of wider objectives, such as the investigation of new products. The guidelines do not cover other types of studies that may involve RWD, such as clinical trials using RWD as a control arm and observational studies.
  2. What is RWD?  The MHRA defines RWD as “data relating to patient health status or delivery of health care collected outside of a clinical study.” RWD may include electronic healthcare records (EHR), primary and secondary care records, disease registries, and administrative data on births and deaths. It may also encompass patient reported outcomes (PRO) data and data which are collected outside of a clinical study setting such as through wearables or other devices.
  3. How could a RCT using a RWD source work in practice? In its simplest form, this may involve patients in a real-world database being randomised to one of a choice of interventions, which may include standard care alone.  Patients are then followed via routine practice. If the trial takes place purely in the real-world setting, no additional data is collected on patients (apart from the data collected in the RWD source) and patients are not blinded to treatment allocation. However, more complex forms are possible. For example, the MHRA discusses introducing blinding of treatment allocation, or implementing hybrid trials if additional endpoints are needed that can't be obtained from RWD sources. In this kind of hybrid model, some of the data are RWD, and some are collected specifically for the trial outside of the RWD source (such as additional clinical assessments or interventions).
  4. Safety as a priority: Of particular importance is the appropriate monitoring of patients in a clinical trial. Patients must be monitored regularly, and mitigation steps introduced depending on the safety profile of the trial and a comprehensive risk-benefit review. Sponsors are legally required to report serious adverse events, as well as suspected unexpected adverse reactions in clinical trials, and these requirements will apply equally to RWD-based trials. The extent of the patient safety monitoring should be determined based on a case-by-case basis, including following an assessment of the risks of the trial intervention(s) relative to standard care and the level of knowledge regarding the medicinal products being tested.
  5. Appropriate endpoints: Endpoints in RWD-based trials must be objective, and ideally, routinely and consistently collected in EHR databases for the patient population in question. Examples of such endpoints include all-cause mortality and inpatient hospitalisations, both of which are well-recorded within the UK general population. It is of key importance that the validity of potential endpoints is not assumed. For example, there is evidence that potential endpoints assumed to be well recorded within the UK, such as the occurrence of acute myocardial infarctions, have in fact been recorded with poor completeness. The MHRA recommends that sponsors conduct feasibility studies prior to the commencement of a RWD trial in order to assess the capture of variables.
  6. A regulatory perspective: Ultimately, the acceptability of RWD-based trials will not be determined by whether the data used is from a real world setting or the result of a traditional RCT. The key is whether the trial is designed to provide the evidence needed to answer the regulatory question at hand. This will be possible where a RCT has been well-designed, as such trials provide high levels of evidence. Where a sponsor is contemplating running a RCT using RWD sources, the MHRA encourages sponsors to engage early with the regulator.

The full guidelines are available here.